Determinants of WTP for Prosecco wine

Abstract

Purpose Under investigation is Prosecco wine, a sparkling white wine from North‐East Italy. Information collection on consumer perceptions is particularly relevant when developing market strategies for wine, especially so when local production and certification of origin play an important role in the wine market of a given district, as in the case at hand. Investigating and characterizing the structure of preference heterogeneity become crucial steps in every successful marketing strategy. The purpose of this paper is to investigate the sources of systematic differences in consumer preferences. Design/methodology/approach The paper explores the effect of inclusion of answers to attitudinal questions in a latent class regression model of stated willingness to pay (WTP) for this specialty wine. These additional variables were included in the membership equations to investigate whether they could be of help in the identification of latent classes. The individual specific WTPs from the sampled respondents were then derived from the best fitting model and examined for consistency. Findings The use of answers to attitudinal question in the latent class regression model is found to improve model fit, thereby helping in the identification of latent classes. The best performing model obtained makes use of both attitudinal scores and socio‐economic covariates identifying five latent classes. A reasonable pattern of differences in WTP for Prosecco between CDO and TGI types were derived from this model. Originality/value The approach appears informative and promising: attitudes emerge as important ancillary indicators of taste differences for specialty wines. This might be of interest per se and of practical use in market segmentation. If future research shows that these variables can be of use in other contexts, it is quite possible that more attitudinal questions will be routinely incorporated in structural latent class hedonic models.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 27 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.